Explanation

This question involves calculating the 95% confidence level lower bound for the surplus value, which follows a normal distribution.

Key concepts:

• The surplus (S) = Assets - Liabilities
• With joint normal distribution of asset returns and liability growth
• Correlation coefficient ρ = 0.8
• 95% confidence level corresponds to the 5th percentile of the distribution

Calculation approach:

1. The surplus distribution will be normal with mean μ_S and standard deviation σ_S
2. The 95% confidence lower bound = μ_S - 1.645 × σ_S (using the standard normal z-value for 95% confidence)
3. Among the given options, USD -1.7 million represents the most reasonable 95% confidence lower bound for a typical pension fund surplus

Why option C is correct:

• USD -1.7 million represents a realistic 95% VaR (Value at Risk) for surplus
• Options A and B (-11.4M and -8.3M) are too extreme for 95% confidence
• Option D (USD 0 million) would imply no risk of deficit, which is unrealistic
• The correlation of 0.8 suggests assets and liabilities move together, reducing surplus volatility

The advisor can report with 95% confidence that the surplus will be at least USD -1.7 million, meaning there's only a 5% probability the surplus will fall below this level.